Statistical Learning by 8-Month-Old Infants - Artigo Científico

Statistical Learning by 8MonthOld Infants Jenny R Saffran Richard N Aslin Elissa L Newport Science New Series Vol 274 No 5294 Dec 13 1996 pp 19261928 Stable URL httplinksjstororgsicisici0036807528199612132933A2743A52943C19263ASLB8I3E20CO3B2Y Science is currently published by American Association for the Advancement of Science Your use of the JSTOR archive indicates your acceptance of JSTORs Terms and Conditions of Use available at httpwwwjstororgabouttermshtml JSTORs Terms and Conditions of Use provides in part that unless you have obtained prior permission you may not download an entire issue of a journal or multiple copies of articles and you may use content in the JSTOR archive only for your personal noncommercial use Please contact the publisher regarding any further use of this work Publisher contact information may be obtained at httpwwwjstororgjournalsaaashtml Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission The JSTOR Archive is a trusted digital repository providing for longterm preservation and access to leading academic journals and scholarly literature from around the world The Archive is supported by libraries scholarly societies publishers and foundations It is an initiative of JSTOR a notforprofit organization with a mission to help the scholarly community take advantage of advances in technology For more information regarding JSTOR please contact supportjstororg httpwwwjstororg Thu Aug 23 191634 2007 infection of nurinecells 15and transgenic mice expressing human CD4 16 and pro vides a rationale for transgenic approaches to developing animal models of HIV disease REFERENCES AND NOTES I F Cocchi et aScience 270 1811 1995 2 Y Feng C C Broder P E Kennedy E A Berger ibid 272 872 1996 3 M Samson 0 Labbe C MollereauG Vassart M Parmentier Biochemistry 35 3362 1996 C J Raport J Gosling V L Schweckart P W Gray I F Charo J 601 Chem 271 17161 1996 4 H Choe et aCell 85 1135 I996 B J Doranzet aibid p 1149 5 T Dragic etaNature 381 667 1996 H Deng et aib p 661GAkhatibetalScience 272 1955 I996 6 S Gartner etaScience 233 215 1986 7 R Atchison eta unpublshedobservatons 8 L Boring et aJ Bioi Chem 271 7551 1996 9 We cloned cDNAs encoding human or murineCCR5 nto the expressionvector pcDNA3 Invitrogenafter engneering the FLAG epitope Into the NHtermnus as descrbed 13Expresson of each construct was determned by FACS wthan antbody to FLAG anti FLAG Boehringer Mannhem and relative expres sion for each see belowwas calculated as the per centage of cells expressng human CCR5 on the cell surface normalizedto the expresson of hCCR5 de fined as loo wthstandard errors of the mean The mean fluorescence intensty of the postve cells from any sngle sample nevervariedfrom the average by more than 30 in a single experimentTherefore neither the relative number of postve cells nor the absolute expression levels wthintransfected cells explains the differences in coreceptor activity C I meric receptors were prepared by the overlap poly merase chain reactlon PCR method 17 hCCR5 HHHH human CCR5 100 relative expression mCCR5 MMMM murlne CCR5 126 2 49 HMMM NHterminus of human CCR5 amnoacids aa 1 to 321 fused to murlne CCR5 aa 35 to 354 77 22 MHHH NHtermnus of murne CCR5 aa 1 to 34 fused to human CCR5 aa 33 to 352 73 t 17MHMM extracellular loop 1 and a por tion of transmembranedomain 3 of humanCCR5 aa 86 to 118 replacing the corresponding segment of the murlne receptor aa 88 to 120 37 t 22 MMHM extracellular loop 2 and adjacent porions of human CCR5 aa 134 to 210 replacing the corre sponding regon of the murine receptor aa 136 to 212 81 2 30 MMHH NHtermnal half of mCCR5 aa 1 to 162 fused to the COOHtermnal half of hCCR5 aa 161 to 352 80 t 39 10 I F Charo et a Proc Nati Acad Sci USA 91 2752 1994 11 C Franci L M Wong J Van Damme P Proost I F Charo J mmuno 154 6511 1995 12 We cloned cDNAs encoding human CCR2B or chi meras into the expresson vector pCMV4 18 after engineeringthe FLAG epitope into the NHterminus as described 13Expressonof each construct see belowwas determned as descrbed earlier Chimer ic receptors were prepared by the overlap PCR method 17 5555 human CCR5 100 relatve expression2222 human CCR2B 87 25222 NHtermnus of CCR5 aa 1 to 32fused to CCR2B aa 45 to 360 27 t 5 2555 NHterminus of CCR2B aa 1 to 44 fused to CCR5 aa 33 to 352 108 172255 CCR2B aa 1 to 136 fused to CCR5 aa 124 to 352 119 t 33 13 F S Monteclaro and I F Charo J Biol Chem 271 19084 1996 F S Monteclaro et ai unpublshed observations 14 J Gosling eta unpubshed observatons 15 P J Maddon eta Cell 47 333 1986 16 P Loreset aAIDS Res Hum Retroviruses82063 1992 17 S N Ho H D Hunt R M Horton J K Pulen L R Pease Gene 77 51 I989 18 S Andersson D L Davis H Dahlback H Jornval D W RussellJ Biol Chem 264 8222 1989 19 M A Godsmth M T Warmerdam R E Atchison M D Miller W C GreeneJ Virol 69 4112 1995 20 COS7 cells were transfected wth2 pg of pasmid DNA per well in a sixwell plate as descrbed 19 DNA samples consisted of appropriate combina tons of 05 pg of a human CD4 expression plasmd pCD4Neo 19l or planvector and 1 5 pg of a chemokine receptorexpressng plasmd or plain vector About 30 hours after additon of DNA the medium in each w e was replaced with 10 ml of medium contanng HVI BaL 1 00 to 170 ng of p24 per sample source N H ADS Reagent Re pository passaged on prlmary human macro phagesAbout 10 hours later an additona 10 ml of medium was added to each well After 30 hours the cells were recovered from the dish as described 19 and analyzed with a FacScan Becton Dickin son Stainincl for intracvtolasmicHV1 w24 was Coulter Immunologyand goat antmouse fuores cein sothocyanate FTCconjugated secondary antbodyBecton DicknsonCells were further stanedwith phycoerythrnPEconjugated antl CD4 Becton DcknsonAppropriate controls n d cated that the appearance of doublepositve cells FTC PE was dependent on cotransfection with both CD4 and human CCR5 expression plasmids and on the presence of HIVI BaL 21 H Arai and I F Charo J 601 Chem 271 21814 I996 22 We acknowledge the advceof M Warmerdam transfectonnfectionassayE Welder FACS stud iesand L Borng H Araand R Speck scientfc interpretation We appreciate the assstanceof J Carrolland M Cenceros In the preparation of ths manuscript Supported In part by NIH grant HL52773 IFCand by Pfzer MAG carredout 7 t hthe Fix and Perm reagents Caltag Laboratories with a monoclonal antibody to p24 24 September 1996 accepted 24 October 1996 Statistical Learning by 8MonthOld Infants Jenny R Saffran Richard N Aslin Elissa L Newport Learners rely on a combination of experienceindependent and experiencedependent mechanisms to extract information from the environment Language acquisition involves both types of mechanisms but most theorists emphasize the relative importance of experienceindependent mechanisms The present study shows that a fundamental task of language acquisition segmentation of words from fluent speech can be accom plished by 8monthold infants based solely on the statistical relationships between neighboring speech sounds Moreover this word segmentation was based on statistical learning from only 2 minutes of exposure suggesting that infants have access to a powerful mechanism for the computation of statistical properties of the language input During early development the speed and accuracy with which an organism extracts environmental information can be ex tremely important for its survival Some species have evolved highly constrained neural mechanisms to ensure that environ mental information is properly interpreted even in the absence of experience with the environment 1 Other species are depen dent on a period of interaction with the environment that clarifies the information to which attention should be directed and the consequences of behaviors guided by that information 2 Depending on the developmental status and the task facing a particular organism both experienceinde pendent and experiencedependent mecha nisms may be involved in the extraction of information and the control of behavior In the domain of language acquisition two facts have supported the interpretation that experienceindependent mechanisms are both necessary and dominant First highly complex forms of language produc tion develop extremely rapidly 3Second the language input available to the young child is both incomplete and sparsely rep Department of Brain and Cognitve Sciences Unversity of Rochester Rochester NY 14627 USA resented compared to the childs eventual linguistic abilities 4 Thus most theories of language acquisition have emphasized the critical role played by experienceinde pendent internal structures over the role of experiencedependent factors 5 It is undeniable that experiencedepen dent nechanisns are also required for the acquisition of language Many aspects of a particular natural language must be ac quired from listening experience For exam ple acquiring the specific words and pho nological structure of a language requires exposure to a significant corpus of language input Moreover long before infants begin to produce their native language they ac quire information about its sound properties 6 Nevertheless given the daunting task of acquiring linguistic information from lis tening experience during early develop ment few theorists have entertained the hypothesis that learning plays a primary role in the acquisition of more complicat ed aspects of language favoring instead experienceindependent mechanisms 7 Young humans are generally viewed as poor learners suggesting that innate fac tors are primarily responsible for the ac quisition of language Here we investigate the nature of the SCIENCE VOL 274 13 DECEMBER 1996 experiencedependent factors involved in language acquisition In particular we ask whether infants are in fact better learners than has previously been assumed thus po tentially reducing the extent to which ex erienceindevendentstructures must be rnaterial that serves as a potential learning experience They are subsequently present ed with two types of test stimuli i items that were contained within the familiariza word stimuli 18 with longer listening times for nonwords Table 1 This noveltv preference or dishabitllation effect ind cates that 8montholds recoenized the dif tion rnaterial and ii items that are highly similar but by some critical criterion were not contained within the familiarization material During a series of test trials that immediately follows familiarization infants control the duration of each test trial by their sustained visual fixation on a blinking light 14 If infants have extracted the crucial information about the familiariza u ference between the novel and the familiar orderings of the threesyllable strings Thus 8monthold infants are capable of extract ing serialorder information after only 2 posited The results demonstrate that in fants uossess owerfulmechanisms suited to learning the types of structures exemplified in linguistic systems Experience may there fore play a more important role in the ac lnin of listening experience Of course simple serialorder informa tion is an insufficient cue to word bound quisition of language than existing theories suggest One task faced by all language learners is the seglnentation of fluent speech into words This process is particularly difficult because word boundaries in fluent seech aries The learner must also be able to ex tract the relative freauencies of cooccur tion items they may show differential du rations of fixation listening during the two types of test trials 15 We used this procedure to determine whether infants can acquire the statistical properties of sound sequences from brief exposures In our first experiment 24 8monthold infants from an AmericanEnglish language environment were familiarized with 2 min of a continuous speech stream consisting of four threesyllable nonsense words hereaf ter words repeated in random order 16 The speech strealn was generated by a speech synthesizer in a monotone fernale voice at a rate of 270 syllables per minute 180 words in total The synthesizer pro vided no acoustic information about word rence of sound pairs where relatively low transitional probabilities signal word boundaries Our next experiment examined whether 8montholds could oerform the are marked inconsistently by discrete acous tic events such as pauses 8 Although it has recently been demonstrated that 8monthold infants can segment words more difficult statistical computations re quired to distinguish words that is recur rent syllable sequences from syllable strings spanning word boundaries that is syllable sequences occurring more rarely To take an English example prettybaby we wanted to see if infants can distinguish a word from fluent speech and subsequently recog nize them when presented in isolation 9 it is not clear what information is used by infants to discover word boundaries This problem is complicated by the variable acoustic structure of speech across different languages suggestiAg that infants must dis cover which if any acoustic cues correlated with word boundaries are relevant to their D internal syllable pair like pretty from a word external syllable pair like tyba Another 24 8monthold infants from an AmericanEnglish language environ ment were familiarized with 2 min of a continuous seechstrealn consisting of native language 10 there is no invariant acoustic cue to word boundaries present in all languages One important source of information that can in principle define word bound boundaries resulting in a continuous strealn of coarticulated consonantvowel syllables with no pauses stress differences or any other acoustic or prosodic cues to word boundaries A sample of the speech strealn is the orthographic string bidakupadotigola bubidaku The onlv cues to word bound u threesyllable nonsense words similar in structure to the artificial language used in our first experiment 19 This time how ever the test items for each infant consisted of two words and two vartwords The partwords were created by joining the final svllable of a word to the first two svllables of aries in any natural language is the statisti cal information contained in seauences of sounds Over a corpus of speech there are measurable statistical regularities that dis aries were the transitional probabilities be tween syllable pairs which were higher within words 10 in all cases for example bida than between words 033 in all cases another word Thus the partwords con tained threesvllable seauences that the in tinguish recurring sound sequences that comprise words from the more accidental fant had heaid duringLfamiliarization but statisticallv over the corvus did not corre spond to words 20 These partwords could onlv be iudped as novel if the infants sound sequences that occur across word boundaries 11 Within a language the transitional probability from one sound to the next will generally be highest when the two sounds follow one another within a word whereas transitional orobabilities for example kupa To assess learning each infant was pre sented with repetitions of one of four three svllable strings on each test trial Two of had learned the words with sufficient spec ificitv and comoleteness that seauences L these threesyllable strings were words from the artificial language presented dur ing familiarization and two were threesyl lable nonwords that contained the same syllables heard during familiarization but crosskg a word boundary were reiatively ilnfamiliar Despite the difficulty of this word versus partword discrimination infants showed a significant testtrial discrimination between the word and partword stimuli 21 with longer listening times for partwords Table 1 TILIS 2 min of exposure to concatenat ed speech organized into words was suffi spanning a word boundary will be relatively low 12 For example glven the sound sequence prettybaby the transtonal prob ability from pre to ty 1s greater than the transitional probability from ty to ba Pre viously we showed that adults and children can use information about transitional not in the order in which they appeared as words 17 The infants showed a significant test trial discrimination between word and non probabilities to discover word boundaries in an artificial language corpus of nonsense words presented as continuous speech with no acoustic cues to word boundaries 13 We asked whether 8monthold infants can extract information about word bound Table 1 Mean tlme spent llstenlng to the famlllar nonwordsand experlment 2 wordsversus patwc times and novel stimuli for experlment 1 words versus 3rdand signlflcance tests comparing the listening Mean listening tlmes s Experiment Matchedpairs t test Famlllar ltems Novel ltems aries solely on the basis of the sequential statistics of concatenated seechWe used the famillarizationpreference procedure de veloped by Jusczyk and Aslin 9 In this 1 797 SE 041 885 SE 045 t23 23 P 004 2 677 SE 044 760 SE 042 t23 24 P 003 procedure infants are exposed to auditory SCIENCE VOL274 13 DECEMBER 1996 1927 cient for 8monthold infants to extract information about the sequential statistics of syllables Moreover this novelty prefer ence cannot be attributed to a total lack of experience with the threesyllable sequenc es forming partwords as was the case with the nonwords in the first experiment Rath er infants succeeded in learning and re membering particular groupings of three syllable stringsthose strings containing higher transitional probabilities surrounded by lower transitional probabilities The infants performance in these stud ies is particularly impressive given the im poverished nature of the familiarization speech stream which contained no pauses intonational patterns or any other cues that in normal speech probabilistically supplement the sequential statistics inher ent in the structure of words Equally im pressive is the fact that 8monthold in fants in both experiments were able to extract information about sequential sta tistics from only 2 min of listening expe rience Although experience with speech in the real world is unlikely to be as concentrated as it was in these studies infants in more nakralsettings presum ably benefit from other types of cues cor related with statistical information Our results raise the intriguing possibil ity that infants possess experiencedepen dent mechanisms that may be powerful enough to support not only word segmen tation but also the acquisition of other as pects of language It remains unclear wheth er the statistical learning we observed is indicative of a mechanism specific to lan guage acqusitionor of a general learning mechanism applicable to a broad range of distributional analyses of environmental in put 22 Regardless the existence of com putational abilities that extract structure so rapidly suggests that it is premature to assert a priori how much of the striking knowl edge base of human infants is primarily a result of expertenceindependent mecha nisms In particular some aspects of early development may turn out to be best char acterized as resulting from innately biased statistical learning mechanisms rather than innate knowledge If this is the case then the massive amount of experience gathered by infants during the first postnatal year may play a far greater role in development than has previously been recognized REFERENCES AND NOTES 1 Certain speciesspecficskllsdevelop wthoutany experiential input ncludngbat echolocatonE Gould Dev Psychobioi 8 33 197511 and cricket song R Hoy Am Zool 14 1067 1974l 2 Examples of behaviors mediated by early experience are imprintngE Hess imprinting Van Nostrand New York 1973M Leon Physioi Behav 14 311 1975l and sucklngresponses n newborn rats M H Techerand E M Blass Science 198 635 1977l 3 These mlestones have been weldocumented both in English for example R Brown A First Language Harvard Univ Press Cambridge MA 197311 and crossingustcally for example E Lenneberg Bio logicai Foundations of Language Wiley New York 1967 D SlobnEd vols 1 to 3 of The Crosslin guistic Study of LanguageAcquisition ErlbaumH s dale NJ 1985 1987 199211 4 This argument from the poverty of the stmulus remains wdelyaccepted for example N Chomsky Aspects of the Theory of Syntax MIT Press Cam bridge MA 1965 S Crain Behav Brain Sci 14 597 199111 5 D BckertonBehav Brain Sci 7 173 19841 N Chomsky Rules and Representations Columbia Univ Press New York 1981J Fodor Modularity of Mind MT Press Cambrdge MA 1983L Gleit man and E Newport in Language An Invitation to Cognitive Science L Gleitman and M Liberman Eds MIT Press Cambridge MA 19951 pp 124 6 Examples ncude vowel structure P K Kuhl K A WlliamsF Lacerda K N Stevens B Lindblom Science 255 606 1992phonotactics P Jusczyk A FriederciJ Wesses V Svenkerud A Jusczyk J Mem Lang 32 401 1993and prosodc struc ture P Jusczyk A Cutler N RedanzChildDev 64 675 199311 7 Exceptions Include research on prenatal exposure to maternal speech A DeCasper JP Lecanuet MC Busnel C GranierDeferre R Maugeais in fant Behav Dev 17 159 1994 and early postna tal preferences J Mehler et a Cognition 29 149 1988l 8 R Cole and J Jakimk n Perception and Production of Fluent Speech R Cole Ed Erlbaum Hlsdale NJ 19801 pp 133163 9 P Jusczyk and R AslnCognitive Psycho1 29 1 I995 10 A ChrstopheE DupouxJ BertonciniJ Meher J Acoust Soc Am 95 1570 1994A Cutler and D Carter Comput Speech Lang 2 133 1987 1IZ Harris Language 31 190 1955 J Hayes and H Clark n Cognition and the Development of Lan guage J Hayes Ed Wiley New York 1970See M Brent and T Cartwright Cognition 61 93 l996l for a dscussion of related statstcal cues to word boundares 12 The transitional probability of frequency of XY IX frequency of x 13 J Saffran E Newport R Aslin J Mem Lang 35 606 19 9 6 R Tunck S BarruecoPsychol Sci n press 14 Each infant was tested ndvidualy whle seated on the parents lap in asoundattenuated booth Syn thetic speech was generated offline by the Macin Talk system and stored on disk at a sampng rate of 22 kHz for online playback through an Audio media board in an Apple Quadra 650 computer An observer outside the testng booth montored the infants looking behavorwththe use of a color video system using a buttonbox connected to the computer to intate trials and score headturn re sponses Both the parent and the observer listened to masking music over headphones to eliminate bias Durngthe 2mn famliarizationphase the infants gaze was frstdirected to a bllnkng Ight located on the front wall of the testngbooth and then the sound sequence was presented from two loudspeakers located on the sdewalls The in fants gaze was drected to one of two blinkng Ights on these side walls during famliarizationbut there was no relatonbetween lights and sound Immediately after famliarization12 test trials were presented six words and SIX nonwords Each test trial began with the central blinking hght When the observer sgnaledwith a button press that the n fant had fixated on the central ight one of the two side blinknglights was turned on and the center light was extnguishedWhen the nfant faced the sde light a head turn of at least 30 n the directon of the Ight the threesyllable test strlng was played and repeated until the infant looked away from the light for 2 s or until 15 s of looking had occurred The observer smply recorded the direc tonof the Infants head turn and the computer measured ookng tmesdetermnedwhen the 2s ookaway crlterlon had been met and controlled the randomization and presentaton of stmul Cu mulatve lookngtme across each of the two types of test trals provded the measure of preference 15 The drectonof the fxatonpreference depends on the degree of famlarity wth the stmuli If the Infants have become hghy famliar wththe stmuli they show dshabituatonbehavior preferrngthe novel stimu 16 Two counterbalanced stimulus condtionswere generated For each condition 45 tokens of each of four trisyllabc nonsense words conditon A tu piro golabu bidaku and padoti condition B dapiku tiiado burobi and pagotu were spoken n random order to create a 2min speech stream wththe stpuaton that the same word never oc curred twce In a row 17 Test stirnull tupiro goiabu dapiku and tilado In condton A the first two strings were words and the last two strings were nonwords the transtonal probablltesbetween the syllables in the nonwords were all zero relatveto the exposure corpus as these syllable palrs had never occurred durng faml iarizatonIn condton B the first two strings were nonwords and the last two strings were words Ths betweensubjects counterbalanced design ensured that any observed preferences for words or non words across both conditions would not be artifacts of any general preferencesfor certain syllablestrings Each of the four test strings were presented repeat ed with a 500ms Interval between test strngs on three differenttrials resutng n a total of 12test trals per Infant 18 There were no signficant differences betweenthe in fants n conditon A and conditon B t22 0 31 The data from the two groups were thus combined for the other analyses 19 ConditonA words pabiku tibudo goiatu and da ropi condtonB words tudaro pigoia bikuti and budopa 20 Test stmulipabiku tibudo tudaro and pigoia In conditon A the frst two strngs were words and the second two strings were partwords For ex ample the partword pigola spanned the word boundary between daropigoiatu and thus was heard durngexposure In condtionB the frst two strlngs were partwords and the second two strlngs were words The partwords were thus threesyllable sequences that the infants had heard during the course of the exposure period The dif fcultyof ths test dscrminaton can be seen by comparing the transtionalprobabilltesbetween the syllables in the words 10 between syllables 1 and 2 and between syllables 2 and 3 to the tran sitonal probabilites between the syllables n the partwords 033between syllables 1 and 2 and 1O between syllables 2 and 3 21 There were no signifcant dfferencesbetween the Infants In conditon A and condtion B 1122 0 49 The data from the two groups were thus combined for the other analyses 22 For examplethis same general mechanismcould be used to find an object such as a human face in the environment 23 We thank J Galipeau J Hooker P Jusczyk A Jusczyk T Mntz K Ruppert and J Sawusch for ther help with varlous aspects of ths research and P Jusczyk S Pollak M SpiveyKnowton and M Tanenhaus for their helpfulcomments on a previous draft Supported by an NSF predoctoral felowshp JRS NSF grant SBR9421064 RNAand NIH grant DC00167 ELN The parents of all particl pants gave Informed consent 10 May 1996 accepted 30 September I996 1928 SCIENCE VOL 274 13 DECEMBER 1996